AI models primarily use a text or speech interface.
Type what you want and it types back. Say what you want it talks back.
This is fancy, a breakthrough, a little showy. And if the user brings the right skills, it’s an extraordinary way to interact.
But the AI UX people (the few that are paying attention, not simply racing to keep up with the engineers) are missing an opportunity.
People prefer multiple choice to essay exams. Go to a restaurant without a menu and people get stressed. They either order something simple or are filled with regret about what could have been.
When the AI prompts us (instead of us prompting the AI), faster progress is possible. When the AI suggests four or five appropriate paths, we’re more likely to consider more options. Building that sort of UX in from the start makes it more likely we’ll expect it.
When all you have is a hammer, everything is a nail. When we design a menu, especially one that changes with context, we get a chance to challenge the user to create variety, possibility and progress.
PS if you’re not using the latest AI models, you’re falling behind. I’m seeing very senior people who are ignoring what’s happening, and the gap is widening. It’s probably worth some time to play with Claude and others.